Scott, if you do not mind, I have prepared a test of Mundorf polyprop x Roederstein polyester.
You are cordially welcome, the thread is provided with measurements as well.
http://www.diyaudio.com/forums/ever...listening-test-preparation-3.html#post3802556
Huh? Antialiasing filters are used at the input of a A/D sampler, to restrict the input signal bandwidth, so that it complies to the sampling theorem requirements.
You probably mean the reconstruction filter (at the output of a D/A), required to remove the HF images of the audio signal.
Firstly can I thank Barleywater for trying all the files and sending me a "sort list", one of the few to actually do this. Thanks. Its much appreciated and the results have been set out in an easily assimilated form.
For everyone else, the clue really is in the title of the thread
Listening Test It was never intended as an exercise in file evaluation and use of software to look for differences. But I know that is what you all resort too. And I wish I was as conversant with all the software as many of you guys are.Can you hear a difference between the two cap choices ? That is all there is to it.
Question... do the files stand up in their own right as a musical experience ? If you played them at home from a bought CD and they sounded as these do, would you be criticising the technical quality on the grounds it was "poor" to listen to or would you think them OK. Would you be happy to use them to evaluate your own gear (assuming you liked the music of course) or would you say they were not worthy. Be honest
Indeed, but this point seems lost to everyone.
Mooly your recording ability just went out the window if that is what is under scrutiny.
Next time you think of such a tests ask for volunteers and send them each a hand full of capacitors, let them figure it out from there.
The door is open and the bar set for anyone wishing to replicate the test with these or similar controls and using files of their own making... files that can, like mine, be put under scrutiny
I think your results are in the post or is that the cheque ! Your observation is spot on though... come on folks, if you can match them then please do so, and if you can't then that is just as valid too.
If your results show you can't match them (but we need the results to know) then you might say, ah... but I know there is a difference... the test is not sensitive enough. Then the door is open for you to come back with a different test that does show a positive statistical result.
We have to start somewhere. Look at the opamp test. Many of you say that's silly and differences are obvious. No one has taken the plunge though and submitted a set of results.
There are no right and wrong answers in this.
That would be great Scott. And if you could give your impressions of the opamp one too
I would suggest everyone trying my tests tries Pavels latest test too. I had a quick listen and sent a result by pm to Pavel.
I suspect that this was done to increase the likelyhood of detection of differences. Baby steps. Make the first test too hard and no one will want to participate in future tests. Make the first test easier and get results that show that the methodology works and perhaps people will be more willing to try in the future with more real world tests.
Tony.
My Hypothesis
Related to my posts #124 and #126 I found that:
In comparing two sound sources, our brain is following two stage processes: (1) Perceiving differences (2) Defining the difference
Stage(1) is accurate, meaning that many people can perceive differences (and it is EASY & ACCURATE!) but it is much MORE DIFFICULT to define this difference (stage(2)). This difference should be defined or labelled by our brain such as "black" versus "not black", "delicious" versus "not delicious", "fatiguing" versus "not fatiguing", "natural" versus "not natural", "tall" versus "not as tall", "beautiful" versus "not so beautiful", etc. Now you can imagine how difficult it is when we are shown a woman with certain beauty and we have to labelled this woman as "beautiful" or "not so beautiful".
Many people have perceived differences in cable sounds, but when tested with DBT they failed miserably. Our conclusion is that they are delusional and there is actually no differences. But after this hypothesis or finding, I tend to believe that they DO hear differences.
By understanding of this hypothesis, we can then find a way to improve the process in stage(2), trying to understand the detail of this process, or knowing how our brain works during this stage, to increase our accuracy in this kind of test.
So, one of the detailed process is our brain to allocate a memory space to store... ah, forget it! My point here is (specific objective of my writing) I started to understand one unique phenomenon during our blind tests. What phenomenon is that? It's a phenomenon where a respondent can get 0/5, 0/6, even 0/10. This is slightly against statistics, don't you think? But can you believe that when there is no difference between the two files, you should not get such result. Statistics says around 50% so 2/6 to 4/6 is fine imo. Interesting? Now you know what to improve in this specific detailed process or stage, no?
Related to my posts #124 and #126 I found that:
In comparing two sound sources, our brain is following two stage processes: (1) Perceiving differences (2) Defining the difference
Stage(1) is accurate, meaning that many people can perceive differences (and it is EASY & ACCURATE!) but it is much MORE DIFFICULT to define this difference (stage(2)). This difference should be defined or labelled by our brain such as "black" versus "not black", "delicious" versus "not delicious", "fatiguing" versus "not fatiguing", "natural" versus "not natural", "tall" versus "not as tall", "beautiful" versus "not so beautiful", etc. Now you can imagine how difficult it is when we are shown a woman with certain beauty and we have to labelled this woman as "beautiful" or "not so beautiful".
Many people have perceived differences in cable sounds, but when tested with DBT they failed miserably. Our conclusion is that they are delusional and there is actually no differences. But after this hypothesis or finding, I tend to believe that they DO hear differences.
By understanding of this hypothesis, we can then find a way to improve the process in stage(2), trying to understand the detail of this process, or knowing how our brain works during this stage, to increase our accuracy in this kind of test.
So, one of the detailed process is our brain to allocate a memory space to store... ah, forget it!
My point here is (specific objective of my writing) I started to understand one unique phenomenon during our blind tests. What phenomenon is that? It's a phenomenon where a respondent can get 0/5, 0/6, even 0/10. This is slightly against statistics, don't you think? But can you believe that when there is no difference between the two files, you should not get such result. Statistics says around 50% so 2/6 to 4/6 is fine imo. Interesting? Now you know what to improve in this specific detailed process or stage, no?
@ tvrgeek: decent PC speakers. I run them loud, so the sound 'washes' over me - that way any nastiness or peculiarities stand out strongly.
Jay, I find a problem with randomly flicking between 2 versions, one slightly better than the other, is that my brain will automatically try and fill in the gaps with the lesser - adjust the poorer cousin so that it matches the better, because my brain will quickly 'learn how it goes', and does quality EQ'ng on the fly, so to speak. I believe the same thing happens when you listen to a favourite song on a really bad car radio - your brain knows what it 'should sound like', and constantly mentally adjusts what's coming in, to match your memory of the song.
The answer is to "listen without listening" - doing it out of the corner of your ear so to speak, take it in as a wash of sound, defocus on whether one's listening to a voice, a violin or piano - it's just a texture of sounds, a pattern of noises.
Jay, I find a problem with randomly flicking between 2 versions, one slightly better than the other, is that my brain will automatically try and fill in the gaps with the lesser - adjust the poorer cousin so that it matches the better, because my brain will quickly 'learn how it goes', and does quality EQ'ng on the fly, so to speak. I believe the same thing happens when you listen to a favourite song on a really bad car radio - your brain knows what it 'should sound like', and constantly mentally adjusts what's coming in, to match your memory of the song.
The answer is to "listen without listening" - doing it out of the corner of your ear so to speak, take it in as a wash of sound, defocus on whether one's listening to a voice, a violin or piano - it's just a texture of sounds, a pattern of noises.
Yeah, that's where we differ. I wouldn't take a refractometer to a taste test.
That would be great. The more results the better
There is an "opamp" test too, and that one needs more participants. Out of the two tests, the active one is the one that should have the most easily audible differences I would have thought.
Of course. The test that has audible differences is more attractive than the one which gives the listener no audible difference. We may need both, I think. To indicate possible audibility of certain component changes. Though we know that such test is not a complete proof, just a kind of indication.
Now, that we know there's a simple file difference analysis, I think there is a good opportunity for another test scheme. People often say that silver is better sounding than aluminum, or boutique cap is better than electrolytic. So here we can test if the person really know the difference or in other words, to tell which is which.
In this Mooly's test I believe that the one with more pronounced treble should be the film cap. I have an idea of technical explanation (I have often posted it) but it is just a guess as I'm not really interested in finding out the scientific explanation. But I think it can be a clue for others who really like to play scientists.
So, which one is which, is it part of the test now (or have been)?
In this Mooly's test I believe that the one with more pronounced treble should be the film cap. I have an idea of technical explanation (I have often posted it) but it is just a guess as I'm not really interested in finding out the scientific explanation. But I think it can be a clue for others who really like to play scientists.
So, which one is which, is it part of the test now (or have been)?
In view of Jay cracking the code as if he
worked at Bletchley Park, and by using file analysis methods the test is now closed. All your "results" were correct Jay, and my initial thought was that it rendered the whole test invalid. Do you know how many hours work went into setting these tests up ?
So Jay extracted the correct key,
file01 = B
file02 = B
file03 = A
file04 = A
file05 = B
file06 = A
file07 = B
file08 = A
file09 = B
file10 = A
file11 = A
file12 = A
However... before I post the results of those who did participate, and they participated in the spirit in which the thread was offered(my thanks, it was appreciated), I will ask another question, the same as Pavel has done in the other test thread.
Which is which. Is A or B the film or the electrolytics. And which do you like best sonically.
Jay ? you did remarkably well in the earlier tests using just your mobile phone to listen to different opamps if I recall correctly. Can you identify the two here ?
worked at Bletchley Park, and by using file analysis methods the test is now closed. All your "results" were correct Jay, and my initial thought was that it rendered the whole test invalid. Do you know how many hours work went into setting these tests up ?
So Jay extracted the correct key,
file01 = B
file02 = B
file03 = A
file04 = A
file05 = B
file06 = A
file07 = B
file08 = A
file09 = B
file10 = A
file11 = A
file12 = A
However... before I post the results of those who did participate, and they participated in the spirit in which the thread was offered(my thanks, it was appreciated), I will ask another question, the same as Pavel has done in the other test thread.
Which is which. Is A or B the film or the electrolytics. And which do you like best sonically.
Jay ? you did remarkably well in the earlier tests using just your mobile phone to listen to different opamps if I recall correctly. Can you identify the two here ?
I personally preffered B, I felt it had more depth than A. I have a sneaking suspicion though that it is the string of electrolytics. It seemed to be a bit richer, perhaps have a little more reverb in it, which could well be added harmonics due to higher distortion of electrolytics.
Jay in future if you find a way of cheating on a test please keep it to yourself, or only privately tell the test organizer. As Mooly said it took a lot of work to set this test up. A lot of other people also spent their time participating. I'm sure that there will be more than one person who is not happy that you have done this.
Although I killed the post within 4 minutes of it being revealed, it nonetheless invalidates the test and all the hard work people had put into it.
Tony.
Jay in future if you find a way of cheating on a test please keep it to yourself, or only privately tell the test organizer. As Mooly said it took a lot of work to set this test up. A lot of other people also spent their time participating. I'm sure that there will be more than one person who is not happy that you have done this.
Although I killed the post within 4 minutes of it being revealed, it nonetheless invalidates the test and all the hard work people had put into it.
Tony.
Last edited:
Tony, Mooly, Pavel,
Sorry for what I have done. The reason for that is because I failed to recognize the ENTERTAINING aspect of this listening test. Since the beginning I have looked at these tests as a SCIENTIFIC approach. If you know what i mean.
This is kind a proof to what I have often mentioned, that for those with good ears, amplitude is NOT the key for preference. Here I prefer A (which has lower gain) for reason that I believe only a few know...
And I believe A is the film cap, based on limited technical knowledge (because I have never heard such cap string).
Well, I only did roughly per chance; yet the differences I was using to sort them by stood out strongly - it's a case of picking the 'right' differences, . No wonder group DBTs always lead to dead ends ...
I'll just repeat the assessment in my PM:
"B substantially better than A, my guess this is the film cap: livelier sound, more detail, sparkle; the harmonics are rendered with conviction - A is dulled, flat in comparison; this possibly would be chosen by people who have 'aggressive' systems; has a hint of the 'tube sound'"
. No wonder group DBTs always lead to dead ends ... I'll just repeat the assessment in my PM:
"B substantially better than A, my guess this is the film cap: livelier sound, more detail, sparkle; the harmonics are rendered with conviction - A is dulled, flat in comparison; this possibly would be chosen by people who have 'aggressive' systems; has a hint of the 'tube sound'"
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.